Toxoplasma gondii is a serious pathogen of humans and livestock in the U.S.A. and world-wide. In addition to its well-known disease-causing abilities in the developing fetus, in the past two decades this protozoan parasite has increased its notoriety through the fatal disease it can cause in AIDS patients. Many studies have shown that the disease outcome in animals and probably people is dependent on which strain of the parasite is responsible for the infection. Knowing the basis for these differences is of great clinical importance in allowing the treatment to be matched to the specifics of the infection. For example, some strains may produce the most serious consequences through an excessive immune response rather than through direct tissue destruction by the parasite itself. Conversely, other strains may be so invasive that the immune response can't keep up and so the patient is overwhelmed by the parasite. Treating these two scenarios requires completely different strategies: in one, steroids or other immunosuppressive treatments might be the best choice while in the other, the immune system may need to be helped, not impaired, and rapid treatment with anti-parasite agents is needed. The goal of our work is to understand the molecular basis for strain-specific differences in the host-pathogen interaction. Our approach has been to use a combination of in vitro and in vivo studies to compare different strains and thereby acquire a detailed understanding of the phenomenology behind the different phenotypes. We use genome-wide analysis of how different strains interact with the infected host cell as well as detailed in vivo analysis that follows the exact progression of disease using methods that allow the infection to be monitored in live animals in a non-invasive way. Once these phenotypes are determined, we use crosses between the different strains of Toxoplasma to map the genes involved. Finally, we employ a combination of molecular genetic and cell biological approaches to verify the involvement of the candidate genes and determine the mechanism of their action with special attention to how the different alleles play out as differences in disease-causing properties. [unreadable] [unreadable] [unreadable]